music-generation

Suno AI music generation and session workflows via Suno-MCP

安装方式

默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

直接命令不会经过审查 Prompt；运行前请先检查来源。

npx skills add https://github.com/sandraschi/suno-mcp --skill music-generation

命令会保持在同一行。复制前请横向滚动并检查完整内容。

想先保存到本地？可下载 SkillsMP 当前能够提供的文件。

决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。

SKILL.md

来源说明 · 只读预览

name	music-generation
description	Suno AI music generation and session workflows via Suno-MCP

Generate AI music from text prompts with optional style and lyrics.
Automate browser sessions: open Suno, log in, generate, download MP3s.
Studio reconnaissance: capture DOM and selectors for future automation (Premier).

Goal	Tools
Start session	`suno_open_browser`, then `suno_login`
Create track	`suno_generate_track`
Download	`suno_download_track`
Status	`suno_get_status`, `get_server_status`
Multi-step (sampling)	`agentic_suno_workflow` with explicit `available_tools`
Studio DOM / cookies	`recon_*` tools

suno_generate_track: prompt (required), style (e.g. synthwave, pop), optional lyrics, duration (auto/short/medium/long).
Credentials: Never commit real passwords; use env or secure local config outside the repo.

Use agentic_suno_workflow with a clear workflow_prompt and tool names such as:

suno_open_browser, suno_login, suno_generate_track, suno_get_status, suno_download_track

Requires FastMCP 3.1+ sampling (client or server-side LLM via SUNO_SAMPLING_*).